REMOTE SENSING OF ATMOSPHERIC TRACE GASES BY OPTICAL CORRELATION SPECTROSCOPY AND LIDAR by Benjamin Thomas Grégory David, Christophe Anselmo, Alain Miffre, Jean-Pierre Cariou and Patrick Rairoux Institute of Light and Matter, Lyon 1 University 1 Lyon 1 University, Institute of Light and Matter, Mirthe Summer Workshop 2014.
38
Embed
REMOTE SENSING OF TRACE GAS BY COMBINING … of a few percents of extinction on weak optical signal ... (AMR) 15 . Water Vapor ... Remote sensing of trace gases with optical correlation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
REMOTE SENSING OF ATMOSPHERIC TRACE GASES BY OPTICAL CORRELATION SPECTROSCOPY
AND LIDAR by
Benjamin Thomas
Grégory David, Christophe Anselmo, Alain Miffre, Jean-Pierre Cariou and Patrick Rairoux
Institute of Light and Matter, Lyon 1 University
1 Lyon 1 University, Institute of Light and Matter, Mirthe Summer Workshop 2014.
Lyon 1 University
2
Study of Trace Gases
Need for local and global spatial distribution of trace gases
Sources and sinks • Localization • Mass flux measurements
Atmospheric model improvements
3
Impact on climate Change the Earth’s radiation budget [IPCC 2007]
Impact on human health
• Lung cancer (World Health Organization) • Harm the cardiopulmonary system • Hazardous gases (methane, benzene, natural gas, hydrogen...)
Gas pipeline in Lybia. Numerical simulation of the
atmosphere dynamics.
Aim of the work
4
The Optical Correlation Spectroscopy Lidar (OCS-lidar)
Goal: Remotely retrieve the concentration profile of a specific trace gas in the atmosphere
Lidar [Fiocco et al., Nature, 1963] [Weitkamp, Springer Ed., 2005]
Correlation Spectroscopy [Sandroni et al., Atm. Env., 1977] [Dakin et al., Sens. Actuators B, 2003] [Lou et al., App. Phys. B, 2009]
A new approach based on:
Inspired by previous works:
Gas correlation spectroscopy lidar [Edner et al., Opt. Lett., 1984] [Minato et al, Jap. J. Appl. Phys., 1999]
Patent [J. Kasparian, J.P. Wolf: FR2916849A1A1]
Outline
5
I. OCS-lidar methodology a. Principle b. Numerical simulations II. First experimental results a. Experimental set-up b. Water vapor measurements III. Conclusion and outlook
Outline
6
I. OCS-lidar methodology a. Principle b. Numerical simulations II. First experimental results a. Experimental set-up b. Water vapor measurements III. Conclusion and outlook
Measurement of a few percents of extinction on weak optical signal (β ≈ 10-7 m-1.sr-1)
Controlling the power density spectrum of the laser pulse (emission, pulse shaping, transmission through the atmosphere)
Advantages
Concentration measurements are sensitive to a specific trace gas (OCS)
Range and time resolved measurements (lidar)
OCS Lidar formalism
8
OCS Lidar Equations :
20
( )( ) ( ) ( ) ( , ) ( , )²i i
K rP r P M r T r dr λ
λ λ β λ λ λ∆
= ⋅ ⋅ ⋅ ⋅ ⋅∫
Using the ratio we obtain a third order polynomial where the unknown is the cumulative concentration CC(r) :
( )( )
C
NC
P rP r
3 23 2 1 0( ) ( ) ( )( ) ( ) ( ) ( ) 0CC r CC rA r A r A r ACC r r⋅ + ⋅ + ⋅ + ≈
With and A3, A2, A1 and A0 depending on
• Measured signals PNC and PC • Laser pulse P0(λ) • Modulator transmission MC(λ) and MNC(λ) • Absorption Cross-Section σ(λ)
0( ) ( ') '
rC rC drC r = ∫ )( ()C r r
rCC∂
=∂
B. Thomas et al., « Remote Sensing of Trace Gases with Optical Correlation Spectroscopy and Lidar », APB, 108, 2012
The OCS-lidar numerical model
9
Study of systematic and statistical errors through the concentration relative error:
input output
input
C CC
−=
B. Thomas et al., « Remote Sensing of Trace Gases with Optical Correlation Spectroscopy and Lidar », APB, 108, 2012
Simulation results for high CH4 concentration
0 500 1000 1500 2000 2500 3000 35000.5
0.6
0.7
0.8
0.9
1.0
PC
PNC
0 500 1000 1500 2000 2500 3000 3500
0
100
200
300
400 Cinput
Couput
OCS
-lida
r sig
nals
(a.u
) Ran
ge c
orr.
Range r (m)
[CH 4]
(ppm
)
10
Parameters: • Methane:
400 ppm
• Wavelength : 1.66 µm • 250 µJ/pulse • 60 000 laser shots • 15 m range resolution
B. Thomas et al., Remote sensing of trace gases with optical correlation spectroscopy and lidar, APB, 2012.
Outline
11
I. OCS-lidar methodology a. Principle b. Numerical simulations II. First experimental results a. Experimental set-up b. Water vapor measurements III. Conclusion and outlook
OCS-lidar experiment
12
Experimental set-up, three main parts :
A femtosecond laser source coupled with an OPA
The amplitude modulation achieved by an Acousto Optical Programmable Dispersif Filter (AOPDF)
The detection system 30 cm diameter Newtonian Telescope Hammatsu photodiode
First experimental proof of the OCS-lidar methodology for water vapor measurements in the visible spectral range.
Active modulation with the AOPDF
13
Acousto Optical Programmable Dispersive Filter [Kaplan et al., Ultrafast Opt. IV, 2004]
Achieve pulse shaping with a ≈ 1 nm spectral resolution.
Based on acousto-optic effect in a birefringent crystal
( )( , ) cose e en z t n n t kzω= + ∆ ⋅ −ne: undisturbed extraordinary refractive index
Δne: amplitude of variation in the extraordinary refractive index.
An acoustic wave (≈ MHz), with angular frequency ω and wavenumber k, generates a refraction index forming a diffraction pattern:
Input Optical beam
TeO2 crystal
Acoustic wave transducer
1425 1450 1475 1500 1525 1550 15750
1000
2000
3000
4000
5000
Powe
r spe
ctra
l den
sity
(a.u
.)
1494 1496 1498 1500 1502 15040
1000
2000
3000
4000
Wavelength (nm)
1.1 nm
4.6 nm
14
Without spectral modulation : MC = MNC = 1
Control measurement 50 µJ/pulse 15 minutes average 35 m spatial resolution 0 500 1000 1500 2000
5000
10000
15000
20000
25000
30000
OC
S-lid
ar s
igna
ls (a
.u) R
ange
cor
r.
PA
PB
0 500 1000 1500 20000.7
0.8
0.9
1.0
1.1
P B/PA
Range (m)
Experimental results for H2O
Set-up with the AOPDF
B. Thomas et al., Remote sensing of atmospheric gases with optical correlation spectroscopy and lidar: first experimental results on water vapor profile measurements, Appl. Phys. B, 2013
First experimental results for H2O (AMR)
15
Water Vapor measurement 50 µJ/pulse 15 minutes average 230 m spatial resolution
0 500 1000 1500 20005000
10000
15000
20000
OCS
-lida
r sig
nals
(a.u
) Ran
ge c
orr.
PNC(r) PC(r)
0 500 1000 1500 20000.7
0.8
0.9
1.0
1.1
P C/PNC
0 500 1000 1500 20000
5000
10000
[H2O
] (pp
m)
Range (m)
Ground concentration [H2O] = 9 200 ppm
B. Thomas et al., Remote sensing of atmospheric gases with optical correlation spectroscopy and lidar: first experimental results on water vapor profile measurements, Appl. Phys. B, 2013
Outline
16
I. OCS-lidar methodology a. Principle b. Numerical simulations II. First experimental results a. Experimental set-up b. Water vapor measurements III. Conclusion and outlook
Conclusion
17
New approach for remote sensing of atmospheric trace gases by coupling Optical Correlation Spectroscopy with lidar (OCS-lidar). Based on a spectrally broadband light source and amplitude modulation.
First experimental proof of the OCS-lidar by measuring water vapor profiles in the atmosphere.
Development of a numerical simulation to study the statistical and systematic errors for methane and water vapor.
Development of a new algorithm to retrieve the trace gas concentration.
B. Thomas et al., Remote sensing of trace gases with optical correlation spectroscopy and Lidar : Theoretical and numerical approach, Appl. Phys B, 108, 689-702, (2012).
B. Thomas et al., Remote sensing of methane with broadband laser and optical correlation spectroscopy on the Q-branch of the 2ν3 band, J. Mol. Spec., Special issue on methane, 291, 3-8, (2013). B. Thomas et al., Remote sensing of atmospheric gases with optical correlation spectroscopy and lidar: first experimental result on water vapor profile measurements, Appl. Phys. B, DOI: 10.1007/s00340-013-5468-4 (2013).
Outlook
18
Validation with standard measurement techniques.
Methane measurement in the infrared spectral range
Field measurements, further investigation on light sources.
Multiple gas monitoring (N2O, CO2, O3, hydrocarbons…) further investigation on the amplitude modulation and other spectral ranges.
B. Thomas et al., « Remote sensing of atmospheric gases with optical correlation spectroscopy and lidar: first experimental result on water vapor profile measurements », APB, April 2013
Statistical error:
25
Dσ
2 2 2N D Bσ σ σ= +
The detector noise
The background noise
Due to :
Bσ
Dσ
Bσ Assess by simulation (Libatran)
Theoretical evaluation:
Experimental evaluation:
Lida
r sig
nal (
mV)
Range (m)
Rela
tive
Freq
uenc
y
Signal (Volt)
σN
Detector Noise Equivalent Power
The signal noise
Systematic error:
26
( ) ( )2 3
0 0 42
0
2 ( ) ( ') ' 2 ( ) ( ') '( , ) 1 2 ( ) ( ') '
2 6
r r
r
TG
C r dr C r drT r C r Odr
σ λ σ λλ σ λ
− ⋅ ⋅ − ⋅ ⋅= − ⋅ ⋅ + + +
∫ ∫∫
0.00 0.05 0.10 0.15 0.20-0.4
-0.3
-0.2
-0.1
0.0
[CH 4]
rela
tive
erro
r
Methane Optical Depth ODCH4
Development of a correction algorithm to reduce the model bias:
45.10output input
input
C CC
−−<
The model bias
Optical Correlation Spectroscopy
27
0.0 0.2 0.4 0.6
Ligh
t swi
tch
Source 1 Source 2
ON
OFF
0.0 0.2 0.4 0.60.98
0.99
1.00
1.01
Dete
ctor
s Si
gnal
(a.u
.)
Time (a.u.)
Input Signal Detector Measurement Signal Detector
a
b
0 10 20 30 40 50 60 70 80 90 1000.00
0.02
0.04
0.06
0.08
0.10
Mod
ulat
ion
fact
or m
(a.u
.)
Cmeas (ppmv)
1 2
1 2
2 I ImI I
−= ⋅ +
Concentration measurement of a target gas in a cell
OCS-Lidar formalism
28 B. Thomas et al., « Remote Sensing of Trace Gases with Optical Correlation Spectroscopy and Lidar », APB, 108, 2012
20
0
20
0
( ) ( , ) ( ) ( ) ( , ) ( , ) ( )( )( )
( ) ( , ) ( ) ( ) ( , ) ( , ) ( )
CC
NNC
NC
K O r P M r T r dP r PP r
K O r P M r T r d
λ λ λ λ β λ λ η λ λ
λ λ λ λ β λ λ η λ λ
∞
∞
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅= +
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅
∫
∫
20
0
20
0
( ) ( ) ( , )( )( )
( ) ( ) ( , )
CC
NCNC
P M T r dP rP r
P M T r d
λ λ λ λ
λ λ λ λ
∞
∞
⋅ ⋅ ⋅=
⋅ ⋅ ⋅
∫
∫
K(λ) and O(r, λ) are achromatic
β(r, λ) = β(r) is assumed to be wavelength independent
η(λ) is assume to be part of the amplitude modulation MC(λ) or MNC(λ)